attach volume failed in redhat6.7 first time afer rebooting system.

[cao biao]

reboot the system and attach volume to instance of os version redhat 6.7 will failed and report error like following:

>>> vop.attach(connection_info)
No handlers could be found for logger "ZVMSDK"
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "volumeop.py", line 463, in attach
    multipath, os_version)
  File "volumeop.py", line 440, in _attach
    raise exception.SDKBaseException(msg=errmsg)
zvmsdk.exception.SDKBaseException: z/VM SDK error: rollback attach because error:SMUT request failed. RequestData: 'cmdVM sample01 CMD 'sudo echo 0x0026000000000000 > /sys/bus/ccw/drivers/zfcp/0.0.1fc5/0x5005076801402797/unit_add'', Results: '{'rs': 1, 'errno': 0, 'strError': '', 'overallRC': 2, 'rc': 8, 'response': ['(Error) ULTVMU0316E On SAMPLE01, command sent through IUCV failed. cmd: sudo echo 0x0026000000000000 > /sys/bus/ccw/drivers/zfcp/0.0.1fc5/0x5005076801402797/unit_add, rc: 8, rs: 1, out: ERROR: ', 'Return code 8, Reason code 1.']}'

But if I execute attach again, the action will succeed.
Reason still not found, open a bug to track this.

[cao biao (flytiger)]

I found the solution but still do not know the reason of this problem.

solution is call config.load_config before init SDKAPI, like this:

from zvmsdk import config
config.load_config()
from zvmsdk import api
sdkapi = api.SDKAPI()
sdkapi.volume_attach(connection_info)

by doing this, the problem will not happen again.
I think this problem will disappear after I add REST API for volume.
Because load_config is called in wsgi now.

[jichenjc (jichenjc)]

so this is a valid bug or not?

[cao biao (flytiger)]

I refine the code and this problem did not happen again. closed.